Designation E 2437 – 05 Standard Practice for Designing and Validating Performance Based Test Methods for the Analysis of Metals, Ores, and Related Materials1 This standard is issued under the fixed d[.]
Trang 1Standard Practice for
Designing and Validating Performance-Based Test Methods
This standard is issued under the fixed designation E 2437; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision A number in parentheses indicates the year of last reapproval A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.
1 Scope
1.1 This practice covers a model for designing and
validat-ing performance-based, ISO 17025-compliant, standard test
methods for the instrumental chemical analysis of metals, ores,
and related materials The principles in this practice can also be
applied to the development of test methods used to determine
the composition of other materials
1.2 This standard does not purport to address all of the
safety concerns, if any, associated with its use It is the
responsibility of the user of this standard to establish
appro-priate safety and health practices and determine the
applica-bility of regulatory limitations prior to use.
2 Referenced Documents
2.1 ASTM Standards:2
E 135 Terminology Relating to Analytical Chemistry for
Metals, Ores, and Related Materials
E 305 Practice for Establishing and Controlling
Spectro-chemical Analytical Curves
E 691 Practice for Conducting an Interlaboratory Study to
Determine the Precision of a Test Method
E 1329 Practice for Verification and Use of Control Charts
in Spectrochemical Analysis
E 1601 Practice for Conducting an Interlaboratory Study to
Evaluate the Performance of an Analytical Method
E 1621 Guide for X-Ray Emission Spectrometric Analysis
E 2027 Practice for Conducting Proficiency Tests in the
Chemical Analysis of Metals, Ores, and Related Materials
E 2165 Practice for Establishing an Uncertainty Budget for
the Chemical Analysis of Metals, Ores, and Related
Materials
2.2 ISO Standards 3
ISO 17025 General Requirements for the Competence of Testing and Calibration Laboratories
ISO Guide 31 Reference Materials—Contents of Certifi-cates and Labels
ISO Guide 32 Calibration in Analytical Chemistry and Use
of Certified Reference Materials
ISO Guide 34 General Requirements for the Competence of Reference Material Producers
3 Terminology
3.1 Definitions—For definitions of terms used in this
prac-tice, refer to TerminologyE 135
3.2 Definitions of Terms Specific to This Standard: 3.2.1 aim interlaboratory uncertainty, n—the maximum
deviation (95 % confidence) to be allowed in the design of the total interlaboratory uncertainty of a test method, beginning with the preparation of a homogeneous sample and ending with
a final report value to the client
3.2.2 aim uncertainty budget, n—during the development of
a standard performance-based test method, the target allocation
of interlaboratory measurement uncertainty among specific components of a measurement process that contribute signifi-cantly to the overall deviation
3.2.2.1 Discussion—The target allocation is made by the
task group as described in Practice E 2165 and serves as guidance for interlaboratory test participants during method testing
3.2.3 interlaboratory uncertainty, n—in a performance
based standard test method, the precision (95 % confidence) that participating laboratories achieved during interlaboratory studies, beginning with the preparation of a homogeneous sample and ending with a final report value to the client
3.2.4 intralaboratory uncertainty, n—in a performance
based standard test method, the precision (95 % confidence) that a laboratory achieves when the method is used by more than one operator In test methods that establish maximum allowable intralaboratory uncertainties, users must be able to demonstrate compliance with those uncertainties in order to report that a given test result was produced using the named method
3.2.5 performance based method, n—a test method that
defines (1) the general approaches for sampling, sample
1 This practice is under the jurisdiction of ASTM Committee E01 on Analytical
Chemistry for Metals, Ores, and Related Materials and is the direct responsibility of
Subcommittee E01.22 on Laboratory Quality.
Current edition approved May 1, 2005 Published July 2005.
2 For referenced ASTM standards, visit the ASTM website, www.astm.org, or
contact ASTM Customer Service at service@astm.org For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on
the ASTM website.
3 Available from American National Standards Institute (ANSI), 25 W 43 rd St.,
4 th Floor, New York, NY 10036 (www.ansi.org).
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.
Trang 2preparation, and making measurements on a specified type of
material, and (2) defines maximum allowable uncertainties for
each measured constituent over its validated concentration
range
3.2.6 uncertainty budget, n—the allocation of
intralabora-tory measurement uncertainty among specific components of a
measurement process that contribute significantly to the overall
deviation
4 Significance and Use
4.1 This practice provides guidance for methods-writing
task group leaders and members in planning, drafting, and
testing performance-based test methods It also guides users of
performance-based methods in interpreting and applying them
properly
4.2 Standard test methods, particularly those that cover the
instrumental analysis of commercial materials by specific
measurement techniques, such as point-to-plane atomic
emis-sion, x-ray fluorescence, inductively-coupled plasma, and
atomic absorption spectroscopy, need sufficient flexibility to
accommodate the various applicable makes and models of
instrumentation At the same time, these standard test methods
must be capable of generating commercially-useful results that
are sufficiently accurate and precise These needs can be met by
the use of performance-based test methods that rely more on
the demonstrated quality of the test results than on strict
adherence to specific procedural steps One model for
devel-oping these methods is described in this practice
4.3 It is expected that laboratories using performance-based
standard test methods developed in accordance with this
practice will prepare their own detailed work instructions
These work instructions will include detailed operating
instruc-tions for the specific laboratory, the specific reference materials
(RMs) employed, and performance acceptance criteria to be
applied in that laboratory It is also expected that those
laboratories will document their own performance data using
their work instruction to show that their data are consistent
with the standard test method’s Precision and Bias statement
Over time, it is also expected that, when applicable, an
individual laboratory’s proficiency test results will also be
consistent with its documented performance
4.4 Traditional ASTM Precision and Bias statements,
devel-oped during interlaboratory testing in accordance with
Prac-tices E 691 and E 1601, provide information on the
perfor-mance achieved by participating laboratories Intralaboratory
precision can be estimated by dividing the Repeatability index,
r, in the Precision and Bias Table by the square root of 2, at any
tested concentration Proficiency test programs, such as those
following PracticeE 2027, periodically provide interlaboratory
performance data from larger numbers of laboratories, using a
variety of test methods, at many different concentrations over
an extended time period Since interlaboratory precision
ob-tained during both method development and proficiency testing
using existing methods is a function of concentration, per
PracticeE 2165, it follows that the interlaboratory testing of a
new test method should perform no worse than its
predeces-sors Practice E 2165describes how historical interlaboratory
test results can be used to set aim data quality objectives for
new test methods This practice incorporates those aim data
quality objectives in developing uncertainty budgets used in the design, development, and testing of performance-based methods This approach ensures that interlaboratory test data included in Precision and Bias statements are consistent with that associated with the other standard test methods
5 Summary of Practice
5.1 The remaining parts of this practice provide instructions for planning a performance-based method development project, conducting the necessary validation and interlabora-tory precision tests, and documenting the results
5.2 The organization of the rest of the practice guides the task group through the following steps:
5.2.1 Section 6—Sets overall objectives for the project, detailed only to the extent needed to convince the task group that the project can be accomplished as planned
5.2.2 Section7—Sets the technical parameters to be incor-porated in the new test method Establishes details needed to facilitate organizing and drafting a test method
5.2.3 Section8—Addresses writing the first draft of the test method
5.2.4 Section9—Describes how to verify the draft method before entering into more extensive interlaboratory testing 5.2.5 Section 10—Discusses writing the interlaboratory protocol plan so that all of the needed data is available in a form that is easily processed
5.2.6 Section11—Describes task group work in conducting the interlaboratory test
5.2.7 Section12—Describes performance-based method in-formation to be included in the Research Report
5.2.8 Section 13—Addresses finalizing the text of the test method after interlaboratory testing is satisfactorily completed
6 Setting Objectives
6.1 For the material to be tested, identify and list applicable national and international product specifications, sampling practices, and standard test methods Be sure to reference these documents appropriately in the new test method
6.2 List the elements and concentration ranges to be in-cluded in the materials to be analyzed using the method and list the subset of those elements and concentration ranges to be incorporated as analytes in the new test method
6.3 Determine how measurement uncertainty and data qual-ity objectives will be handled
6.3.1 To comply with ASTM requirements, plan to prepare
a precision and bias statement To comply with ISO 17025
requirements, plan to identify all significant sources of mea-surement uncertainty and to calculate the intralaboratory un-certainty (3.2.3) from the information in the precision and bias statement
6.3.2 To help laboratories use the new performance-based test method more effectively, the task group should consider developing an uncertainty budget (3.2.5) That budget may be developed as suggested in this practice, or in a different way, provided that the minimum requirements in6.3.1are met 6.3.3 If the task group elects to develop an uncertainty budget as described in this practice, it should attempt to establish each component of its aim uncertainty budget as defined in Practice E 2165 In some cases, the method of
Trang 3choice may not be optimum for each analyte to be covered In
those cases, if the anticipated performance meets applicable
specification requirements, the task group should set its data
quality objectives in a way that is compliant with specifications
(commercial needs) and attainable performance The
compo-nents of uncertainty will be useful in planning and developing
the performance-based test method
7 Setting Technical Parameters
7.1 Select the Approach to Make the Measurements:
7.1.1 Select a measurement technique that represents an
optimum choice based on: (1) wide acceptance and use
throughout the industry, (2) anticipated ability to measure all
elements and concentration ranges in conformance with the
aim uncertainty budget, (3) anticipated sustained availability of
Certified Reference Materials (CRMs) for achieving traceable
calibrations (Note 1) of all elements and concentration ranges,
and (4) anticipated sustained availability of homogeneous
materials needed for statistical control of the method
N OTE 1—This practice assumes the use of a Type III calibration as
defined in ISO Guide 32 Types I and II do not require the use of CRMs.
7.1.2 Select the sampling equipment and procedure
Re-solve any questions about the adequacy of the sampling
process Ensure its acceptability to users If a new or revised
sampling practice is needed, arrange for the development of
that practice and establish acceptability with those who are
responsible for performing the sampling
7.1.3 Select a sample preparation technique that is
compat-ible with the selected sampling and measurement technique
and current industry practices If a new or revised sample
preparation practice is needed, arrange for the development of
that practice as part of the new test method
7.2 Design the Measurement Approach:
7.2.1 Calibration Approach—Based on the measurement
technique selected in 7.1.1, determine if calibration is to be
achieved by matrix-matched CRMs or by laboratory-prepared
RMs that are traceable to CRMs Matrix-matched CRMs are
preferred, when available, and are most frequently used when
solid samples are analyzed RMs that are traceable to CRMs
are often used in test methods calibrated with solutions In
these cases, materials of known purity are added to CRM
solutions in order to match the overall composition of sample
solutions Refer toISO Guide 32for recommended approaches
to calibration using Certified Reference Materials
7.2.2 Data Quality Objectives—Set the aim intralaboratory
uncertainty (3.2.2) to be achieved in the new test method
Complete the aim uncertainty budget for all elements and
concentration ranges in accordance with the decisions made in
Section6 Refer toX1.2and PracticeE 2165for one approach
7.2.3 Calibration and Calibrants:
7.2.3.1 For each element and concentration range, ensure
the availability of sufficient numbers of acceptable calibrants
Sufficient numbers of calibrants must be available to define the
shape of each calibration curve over the concentration range of
interest and to allow compensation for interelement effects
Refer to Practice E 305 or Guide E 1621 for additional
background information An acceptable calibrant is a CRM that
is compliant withISO Guide 34, or an RM that is traceable to
a CRM or the appropriate SI unit, and has an estimated uncertainty small enough to make it possible to meet the aim data quality objectives of the method Some methods may be calibrated directly with CRMs used in the as-received condi-tion Others may require the dissolution of chip or solid CRMs
or the use of CRM spectrometric solutions that must be combined with other reagents or materials in order to provide suitable calibrants In cases where CRMs are combined with other reagents to make calibrants, the standard test method must provide strict specifications on those reagents to ensure that the calibrants’ assigned quantity values are accurate and have correct uncertainty estimates
7.2.3.2 Set aim uncertainty limits for all calibration curves and calibrants See X1.2.2.2 and X1.2.2.3 for a suggested approach
7.2.4 Verifiers—Ensure the availability of traceable
refer-ence materials that comply with the requirements of7.2.3, but are reserved for use to independently verify the calibration function
7.2.5 Quality Control—For each element and concentration
range to be determined, ensure the availability of material that can be used for control in accordance withISO 17025, clause 5.9 For planning purposes, consider the availability of at least two concentrations for each calibration range, one at about
20 % and one at about 80 % of the maximum values Avail-ability means that a typical laboratory can acquire the needed materials in a reasonable way, either commercially or in-house Establish aim uncertainty limits for each case SeeX1.2.2.1for
a suggested approach
8 Drafting the Method
8.1 Draft the standard test method in accordance with the ASTM Form and Style Manual, Part A, making sure to include the following information which is essential to a performance-based method Except for the intralaboratory performance requirement, the draft will include estimated uncertainty infor-mation consistent with the decisions made in Section6
8.2 Significance and Use—Emphasize that this is a
performance-based method and that each user is expected to create specific work instructions that describe how the performance-based method is applied in that laboratory Also, emphasize that the user laboratory is expected to have perfor-mance data taken in accordance with its work instructions to demonstrate that it meets the minimum data quality objectives specified in the standard test method
8.3 Interferences—In a performance-based method, it is not
practical to identify all possible interferences that may cause bias in the test results In drafting this section of the test method, the task group may identify common interferences and advise laboratories to take steps to avoid them Likewise, this section shall include a statement to the effect that the user laboratory is responsible for ensuring the absence of, or correcting for, interferences that may bias test results generated while following its specific work instructions
8.4 Apparatus—Identify the generic types of equipment
covered in the test method For example, “sequential and simultaneous X-ray fluorescence spectrometers capable of covering the elements and concentration ranges included in
Trang 4laboratory’s documentation and capable of meeting the
perfor-mance requirements specified in the method.” Also, include the
functionality required of sampling and sample preparation
equipment, as appropriate, taking care not to specify more
prescriptively than necessary to remain within the scope of the
method
8.5 Reference Materials, Reagents, and Related Materials.
8.5.1 Calibrants:
8.5.1.1 List the requirements for calibrants and give
instruc-tions for selecting them When defining the requirements for
calibrants, refer to the decisions made in 7.2, while trying to
avoid, if possible, specifying the use of specific CRMs
8.5.1.2 For a comparative method to be calibrated using
CRMs, identify the qualities needed in the calibrants, such as:
“Select sufficient numbers of certified reference materials that
are supplied by National Metrology Institutes or are certified to
comply with ISO Guides 31 and 34 Cover all elements and
concentration ranges and allow for the correction of
interfer-ences, as needed.” The standard test method should not specify
which specific CRMs to use because the test method will
become invalid when those specific CRMs no longer exist
Similarly, new CRMs that become available after the method is
published would not be included If, for some reason the use of
a specific CRM is unavoidable, include language that allows
the use of its replacement
8.5.1.3 For a method to be calibrated using solutions or pure
chemicals, describe how to obtain or prepare calibrants that
meet the traceability requirements without creating bias by the
addition or loss of analyte concentration
8.5.1.4 If the task group elects to develop a full uncertainty
budget in accordance with Appendix X1, or an alternative
approach, refer to the maximum aim uncertainty to be allowed
for the calibrants as a function of concentration
8.5.2 Verifiers—Specify the requirements for verifiers that
meet the requirements of8.5.1but specify that they not be used
as calibrants Require that verifiers be analyzed as unknowns
immediately after the calibration is complete Since verifiers
are used to confirm the accuracy of the calibration process, the
limits of uncertainty for the verifiers and the verification
process are the same as for the calibrants and the calibration
process
8.5.3 Standardization Materials—If the method requires
standardization (drift correction), give requirements for the
selection and management of the materials Consider the effect
of standardization on the final report value and establish
criteria that are consistent with the acceptable intralaboratory
precision It is not necessary to establish specific limits of
acceptability for the standardization procedure because these
sources of variation are included in the control process
8.5.4 Quality Control— Specify the requirements for
estab-lishing the number and concentration of control samples, as
needed Concentrations need not be certified, but homogeneity
should be established and should be about the same level as the
calibrants If the use of control charts is to be specified, refer to
Practice E 1329
8.5.5 Reagents:
8.5.5.1 Water—If water is used in the test method and the
purity of the water might influence the quality of the test
results, specify the purity requirements Pay particular attention
to setting analyte concentrations that, if exceeded, and not corrected for, could cause bias in the test results
8.5.5.2 Chemicals—If specific chemicals, liquids, solids, or
gases, are to be used in the test method, list them and their required purities If the laboratory has the option to select its own chemicals, give generally applicable specifications If no chemicals are to be used in the test method, skip this section
8.6 Procedure:
8.6.1 Sampling of lots of materials, whether conducted by laboratory-supervised personnel or not, is usually considered outside the scope of an ASTM test method However, for some materials, laboratories may be required to subdivide the as-received sample in order to perform the required tests Such sub-divisions should be considered and described in the test method
8.6.2 Indicate the major steps that must be accomplished to generate a report value Refer to manufacturer’s instructions and individual user’s work instructions for details
8.6.3 Discuss and explain the criteria of acceptance that must be met if a report value is to be supplied in association with the Standard Test Method See X1.2.1 for a suggested approach State the intralaboratory data quality objectives plus any measurement uncertainty requirements that must be met and documented by the laboratory, whether reported to the client or not
8.7 Precision and Bias:
8.7.1 Write this section after completion of the interlabora-tory test program, as required by standard ASTM protocols, including PracticesE 1601andE 691
8.7.2 Add a section that summarizes the derivation of the intralaboratory test data and any detailed measurement uncer-tainty budget requirements the task group elected to include in the method Use Appendices and Annexes as appropriate
8.8 Report:
8.8.1 Require that a laboratory comply with all require-ments, including data quality of the standard test method in order to state on a test report that the test result was generated using this test method Require the laboratory to comply with all reporting requirements of ISO 17025
8.9 Annex or Appendices, or both:
8.9.1 Provide sufficient detail on how the uncertainty budget data was generated during the interlaboratory study and how the requirements summarized in the Precision and Bias state-ment shall be interpreted by laboratories that use the standard method Use either an Annex or an Appendix, as appropriate
9 Verifying the Drafted Method
9.1 Verification Laboratory—Select a competent laboratory
(one that complies with the applicable clauses ofISO 17025) which utilizes the measurement technique to analyze the material to be covered in the new standard test method Obtain sufficient performance data from that laboratory to demonstrate that the technique of choice can achieve the expected data quality objectives for all elements and concentration ranges, making sure to cover any which might be difficult to achieve
If the results from the selected laboratory cast doubt on the ability to achieve ultimate success, consider redesigning the proposed test method This step is primarily intended to
Trang 5provide design information to help the task group effectively
plan its work program Therefore, design details are left to the
task group
9.2 If the test materials and interlaboratory test protocol are
available as described in Section10, the task group may elect
to have the verifying laboratory perform the interlaboratory
study protocol as a test case If successful, the results of the
verifying laboratory may be used as a participating laboratory
in the final calculations
10 Drafting the Interlaboratory Test Protocol
10.1 Select Test Materials—Select test materials that cover
the elements and concentrations to be included in the new
method Make sure that suspected interferences are also
covered To the fullest extent possible, use CRMs that have
reliable uncertainty estimates available so as to be able to
demonstrate absence of measurable and correctable bias If the
test method includes sample preparation steps that might
influence the quality of the results, be sure to include some test
materials that need to be prepared by the participating
labora-tories and be prepared to evaluate the precision obtained on
these materials This will allow the task group to evaluate the
effects of sample preparation on the precision Be sure that the
homogeneity of all test materials is known, including CRMs
10.2 Introduction:
10.2.1 Explain that this test protocol covers the
develop-ment of a performance-based method and that the performance
of participating laboratories will be used in establishing the
data quality that all laboratories that will use the new method
will need to achieve in order to claim that they followed the test
method Explain that the Interlaboratory Study is based on Plan
A in PracticeE 1601and that all test materials are known to be
homogeneous
10.2.2 Explain that the aim data quality objectives included
in this work plan have been achieved by competent laboratories
using similar test methods over time and by at least one
competent testing laboratory using the draft method under test
10.2.3 Advise the participating laboratory that it is expected
that the aim data quality objectives should be met during the
formal interlaboratory test If the participating laboratory
cannot meet the aim data quality objectives, it should check the
function of its equipment, and if that fails, it should contact the
task group chairman immediately
10.3 Qualification of Participating Laboratories:
10.3.1 In order to avoid receiving sub-par data, it is
sug-gested that the task group ask the participating laboratory to
provide data that shows that it is capable of obtaining
accept-able data prior to actually conducting the test The task group
shall design the qualification test and request that the
qualifi-cation data be generated before conducting the interlaboratory
test The qualification data shall be submitted to the
coordina-tor no later than with the final data package
10.3.2 Typical qualification packages might include
demon-stration of ability to achieve adequate signal/noise ratios,
demonstrated ability to calibrate over the concentration ranges
of interest, and sufficient measurement precision over the
anticipated concentration range to achieve the desired
analyti-cal performance In some cases these requirements might be
fulfilled by asking the participating laboratory to analyze
specified CRMs a given number of times, back to back, on one day, using the participating laboratory’s existing in-house procedure
10.3.3 In designing the qualification and final tests, the task group should be careful to give specific directions so that a laboratory will know immediately whether or not its measure-ment performance meets expectations In the event that the laboratory does not meet expectations, it has time to resolve the problem before spending time and resources on the official test
10.4 Instructions for Carrying Out the Interlaboratory
Study:
10.4.1 The instructions for conducting the interlaboratory test should be limited to the test itself and should not alter any instructions in the draft test method Give instructions to assist the task group with evaluation of data, for example, by providing data sheets Data sheets should make clear the number of significant figures to be supplied by participating laboratories Without such instructions, the task group may receive data in a form that is not optimum for statistical analysis Refer to Practice E 1601 for further instructions regarding statistical calculations
10.4.2 The following types of information might be rou-tinely required by the task group: (1) all pre-qualification data, (2) make and model of sample preparation and measuring equipment, the use of which might affect data quality, (3) list of CRMs used for calibration, with copies of certificates for each showing measurement uncertainties, (4) Calibration records showing the degree of curve fit and calibration measurement uncertainty, (5) list of RMs used for control and standardiza-tion, as appropriate, with documentation showing their homo-geneity, (6) all test data used to calculate each report value included in the interlaboratory test, and (7) statements relating
to compliance with aim quality objectives
10.4.3 Provide information on the measurement uncertainty expectations being placed on the cooperating laboratories and provide resources to contact with questions
10.4.4 Ask for comments from the participating laboratory 10.4.5 Set a deadline for submission of all test results to the task group chair
11 Conducting the Interlaboratory Test
11.1 For each participating laboratory, prepare a packet containing: (1) a cover letter (optional, but recommended), (2) the drafted method, (3) instructions for the interlaboratory test, (4) report forms, including data sheets, and (5) test samples Refer to Practice E 1601 for background on conducting an interlaboratory test
11.2 Receive data sets and comments from all participating laboratories
11.2.1 Review the qualification data from each laboratory to ensure that the laboratory was able to meet the general competency requirements associated with the test If the qualification data is not acceptable, contact the laboratory and resolve the issues before evaluating the test method data The availability of qualification data helps ensure that measurement uncertainties calculated as a result of the interlaboratory test are representative of competent laboratories If the qualifica-tion data is satisfactory, but the quality of the test data is not, there may be good reason to question the draft test method
Trang 611.3 Prepare a Precision and Bias table and statement in
accordance with ASTM requirements Refer to Practices
E 1601 or E 691 Add a brief statement that defines the
maximum intralaboratory precision that a laboratory can
achieve and still be able to report results in accordance with the
new standard test method
11.4 If the task group elected to include a measurement
uncertainty budget as part of the test method, add an additional
section to the Precision and Bias statement that describes the
maximum uncertainty that a laboratory can achieve at each step
and refer to the appropriate appendix/annex for details
12 Writing the Research Report
12.1 Prepare the Research Report, covering all ASTM
requirements and discussing fully the derivation of
measure-ment uncertainty data, as described below
12.2 Summarize the pre-qualification data so that the official
record of the test method will contain evidence that the
participating laboratories were capable of performing the test
method Exclude the data from any non-compliant laboratories
from further evaluation
12.3 For each element in each test sample, calculate the
parameters required in Practice E 1601 Prepare the usual
Precision and Bias Table
12.4 For each element on each test sample, calculate the
interlaboratory precision, 95 % confidence, as described in
PracticeE 1601 Divide by the square root of 2 to estimate the
intralaboratory precision Compare the intralaboratory
preci-sion values with the aim data quality objectives If acceptable,
draw a best fit line through the points on a log-log plot, and create a smoothed data table for each element, or, preferably, for all elements, if appropriate This relationship defines the required intralaboratory performance (95 % confidence), that must be demonstrated in order to report that a set of test results were generated using the new standard test method
12.5 If the task group wrote the test method including an uncertainty budget, perform calculations as described in 6.3.3
for control, calibration, and reference materials
13 Finalizing the Method
13.1 Review and revise the draft method as needed for accuracy and completeness according to the usual ASTM procedures as supplemented by this practice
13.2 Provide the Precision and Bias section Add a section
to the Precision and Bias section defining the intralaboratory performance requirements as calculated in 6.3.3 Add a sen-tence indicating that a laboratory that uses this method must meet these performance criteria in order to claim that a report was generated using this test method
13.3 If an uncertainty budget is included in the method, provide a summary of those requirements either as a separate section or as an appendix or annex, as appropriate One acceptable way to present the numerical data is in the form of
a Table, similar to that shown in PracticeE 2165
14 Keywords
14.1 analytical chemistry; measurement uncertainty; mea-surement uncertainty budget; performance-based test method
APPENDIX (Nonmandatory Information) X1 Suggested Means for Establishing Aim Measurement Uncertainties for Performance-Based Methods and Confirming Them
During Interlaboratory Testing X1.1 Introduction
X1.1.1 As described throughout this practice,
performance-based methods rely much more heavily on the quality of the
test results than on obedience to prescriptive experimental
procedures It follows that those who write performance-based
standard test methods must be able to clearly define acceptable
performance This practice defines acceptable performance as
the intralaboratory precision (95 % confidence) obtained
dur-ing interlaboratory testdur-ing It then provides one approach for
task groups to incorporate model uncertainty budgets in
standard test methods, also in harmony with PracticeE 2165
X1.1.2 A primary advantage associated with the use of this
uncertainty practice in conjunction with the aim uncertainties
given in PracticeE 2165is that the task group will be assured
that its aim uncertainty objectives are comparable to other
standard test methods and to proficiency test performance in
general This means that laboratories that participate in the
interlaboratory testing can be reasonably confident that the aim
uncertainties set by the task group are reasonable and
achiev-able It also means that test results achieved by the laboratories
that use the test method will comply with general good laboratory practice and that the laboratory’s results submitted
to proficiency test programs will be consistent with other participants and other test methods
X1.1.3 If the task group sets data quality objectives that are less stringent than those identified in Practice E 2165, then there is a lower probability that laboratories that use the method and contribute to proficiency tests that include other test methods, will perform as predicted by the model
X1.2 A Model for Establishing Aim Data Quality Objectives
X1.2.1 Intralaboratory Precision—The task group should
establish acceptability criteria for intralaboratory precision as guidance for the cooperators in the interlaboratory test For any concentration, this value can be calculated as described in Practice E 2165 Be sure to report the results as 95 % confi-dence in order to comply with ISO 17025
X1.2.2 Uncertainty Budget—It has been shown that the
performance (precision without bias) obtainable by competent
Trang 7laboratories performing optimized, state-of-the-art methods
can be described by a straight line on a log-log plot of
performance vs concentration4This model has been verified
using both proficiency test data and interlaboratory testing of
new standard methods of analysis A scheme for applying these
principles to uncertainty budgets is provided in Practice
E 2165 This document applies those principles to writing
performance-based standard test methods
X1.2.2.1 Control—Divide the intralaboratory precision by
the square root of 2 to find the aim uncertainty for the use of
control materials when performing the test method Note that
the aim control uncertainty is expressed as 2 sigma or 95 %
confidence It may be helpful to interpret control charts using
the Westgard Rules as described in PracticeE 1329
X1.2.2.2 Calibration—Divide the aim uncertainty for the
control function by the square root of 2 to find the aim
uncertainty for the calibration function This may be
inter-preted as the maximum difference between the assumed true
value of any calibrant and the calculated curve fit through that
point For purposes of assessing the calibration uncertainty, the
task group may assume that, because the curve fit is a
calculated function that averages responses from several
ref-erence materials, all points along the calibration curve should comply with the 95 % confidence
X1.2.2.3 Calibrants—Divide the aim uncertainty for the
calibration function by the square root of 2 to find the aim uncertainty at the assumed true concentration value for each calibrant
X1.3 Confirming the Measurement Uncertainty Budget during Interlaboratory Testing
X1.3.1 If the task group has determined that a full uncer-tainty budget is to be included in the standard test method, each aim uncertainty budget item shall be included in the Interlabo-ratory Study work plan so that each laboInterlabo-ratory knows that it is expected to meet or exceed these uncertainties during its testing of the method
X1.3.2 When all test results are in, the task group will compare the aim uncertainties with the experimentally deter-mined values and then adjust the uncertainty budget values as necessary However, if the experimentally obtained uncertain-ties are significantly worse than the aim values (confirmed during verification testing), there is reason to believe that the new test method may not be fully optimized and that further revision is needed
X1.3.3 The data, the findings, and an explanation of deci-sions made shall be included in the Research Report A summary shall be included in the test methods for use by laboratories in implementing the new test method
ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.
This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.
This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org).
4 D A Flinchbaugh, L F Crawford, and D Bradley, “A Model to Set
Measurement Quality Objectives and to Establish Measurement Uncertainty
Expec-tations in Analytical Chemistry Laboratories Using ASTM Proficiency Test Data,”
Accreditation and Quality Assurance, (2001) 6:493-500.